基于显式和隐式社交网络的混合推荐  

MIXED RECOMMENDATION BASED ON EXPLICIT AND IMPLICIT SOCIAL NETWORKS

作　　者：王帅[1] 兰少华[1] 

机构地区：[1]南京理工大学计算机科学与工程学院,江苏南京210094

出　　处：《计算机应用与软件》2016年第11期37-40,48,共5页Computer Applications and Software

基　　金：国家自然科学基金项目(61170035)

摘　　要：传统的社交网络推荐一般依靠用户之间的好友关系,但好友关系不是基于共同兴趣而产生的。针对这种情况,提出通过用户标签所表达的情感兴趣来扩展用户好友关系,形成基于用户好友关系和共同兴趣的混合推荐。利用用户间直接的朋友关系构建显式社交网络,利用标签数据构建隐式社交网络;在显式和隐式社交网络图中分别采用提出的SNA_SPFA(Social Networks Algorithm Based on Shortest Path Faster Algorithm)算法得到推荐结果;最后按照一定权重混合两种推荐结果。实验表明,该方法优于传统的协同过滤方法和社交网络推荐。Traditional social networks recommendation usually relies on friendships between users,but the friendships are not based on common interests. In light of this situation,we propose to expand users' friendships by emotion and interest expressed in users' tags,and to form the mixed recommendation based on users' friendships and common interests. First,we use direct friendship between users to construct an explicit social network,and use tag data to construct an implicit social network. Then we apply the proposed SNA_SPFA algorithm to explicit and implicit social graphs respectively to get recommendation result. Finally,we mix the two recommendation results according to a certain weight.Experiments show that this method is superior to traditional collaborative filtering methods and social network recommendations.

关 键 词：社交网络 标签 情感 混合推荐 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]